Create an index from a string:

Create an index from a file:

Create an index from a URL:

Create an index with a specific name:

Annotate sources with a label:

Each chunk inherits the correspondent source label:

The label is returned when performing search:

Annotated sources with tagged metadata:

Specify the annotations in a separate Association:

Specify a custom distance function for the index:

By default, EuclideanDistance is used:

Train a custom feature extractor:

Use it to extract features from another text:

Specify a custom location to store the index:

Retrieve the location:

By default, the index is stored in a local object:

Store the vector index in a file:

Retrieve the location:

Recreate the database from the file reference:

Create a text with multiple very short entries:

The entire text will be embedded as a single chunk:

Adjusting the minimum and maximum item length to chunk into more relevant sections:

Create several paragraphs of text:

Join them with nonstandard separators:

The default automatic paragraph and sentence chunking will give poor results:

Use a custom split pattern to create chunks at the custom dividers:

The index's automatic location is determined by its name:

With default OverwriteTargetAutomatic, a new index name is generated to avoid collisions:

To force overwriting, use OverwriteTargetTrue:

Use OverwriteTargetFalse to perform a strict check:

OverwriteTargetFalse will also prevent reusing the same index name in a different location:

Create a file:

By default, existing files are not overwritten:

Use OverwriteTargetTrue to overwrite the existing file:

Specify a custom working precision for the embedding vectors:

The working precision is stored in the index's vector database:

Retrieve the precision value: